CN106767788A  A kind of Combinated navigation method and system  Google Patents
A kind of Combinated navigation method and system Download PDFInfo
 Publication number
 CN106767788A CN106767788A CN201710005473.4A CN201710005473A CN106767788A CN 106767788 A CN106767788 A CN 106767788A CN 201710005473 A CN201710005473 A CN 201710005473A CN 106767788 A CN106767788 A CN 106767788A
 Authority
 CN
 China
 Prior art keywords
 navigation system
 data
 system
 navigation
 estimate
 Prior art date
Links
Classifications

 G—PHYSICS
 G01—MEASURING; TESTING
 G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
 G01C21/00—Navigation; Navigational instruments not provided for in preceding groups G01C1/00G01C19/00
 G01C21/10—Navigation; Navigational instruments not provided for in preceding groups G01C1/00G01C19/00 by using measurements of speed or acceleration
 G01C21/12—Navigation; Navigational instruments not provided for in preceding groups G01C1/00G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
 G01C21/16—Navigation; Navigational instruments not provided for in preceding groups G01C1/00G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
 G01C21/165—Navigation; Navigational instruments not provided for in preceding groups G01C1/00G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with noninertial navigation instruments

 G—PHYSICS
 G01—MEASURING; TESTING
 G01S—RADIO DIRECTIONFINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCEDETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
 G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
 G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
 G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting timestamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
 G01S19/42—Determining position
 G01S19/45—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement
 G01S19/47—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement the supplementary measurement being an inertial measurement, e.g. tightly coupled inertial
Abstract
Description
Technical field
The present invention relates to field of navigation technology, more particularly to a kind of Combinated navigation method and system.
Background technology
In current state of the art, there are satellite navigation, inertial navigation, earthmagnetism navigation, terrain match navigation etc. various to lead Boat mode.It is because inertial navigation is entirely autonomous navigation mode and relatively continuous with information, interference is not easily susceptible to, it is short The advantages of precision is higher in time, therefore, inertial navigation still occupies an important position in current all navigation modes.It is prompt Inertial navigation (SINS) system is the one kind in inertial navigation, the basic characteristics with inertial navigation.But, strap down inertial navigation is led Boat is a kind of calculating of stepping type, and its error will be increased over time and increased sharply, it is impossible to which satisfaction keeps certain for a long time The requirement of navigation accuracy.
The appearance of integrated navigation solves the subproblem that current single navigation mode runs into.Integrated navigation be by two kinds or The two or more navigation mode of person is used in combination with, and reaches the purpose learnt from other's strong points to offset one's weaknesses.By taking most common satellite navigation as an example, defend Star navigation has the advantages that roundtheclock, effective in larger spatial dimension, navigation accuracy is higher, error is not with time integral. But compared with strapdown inertial navigation system, satellite navigation system also has that information updating frequency is relatively low, hold under the conditions of Larger Dynamic The shortcomings of easy losing lock, signal are easily disturbed.And strapdown inertial and satellite navigation are constituted into integrated navigation system, then can be with Realize highprecision independent navigation for a long time.
Integrated navigation system needs to use information fusion technology.Current fusion method have Kalman filtering, robust filtering, The various ways such as particle filter.Kalman filtering is a kind of Linear Minimum Variance Filter, for the line under white noise drive condition Sexual system, its filter value is optimal.But, Kalman filtering needs system noise covariance battle array and observation noise covariance matrix To describe the precision of time renewal and the precision of observation respectively.However, system noise covariance battle array of the prior art and sight It is all rule of thumb to set to survey noise covariance battle array, therefore causes the setting of variance matrix not accurate enough rationally, so as to influence filtering Precision.
The pine combination mode of strapdown inertial and satellite navigation is generally used in the prior art, and the method is right The variables such as the velocity location of inertial navigation are corrected, and do not consider the constant value drift and ratio of gyro and accelerometer The change equal error source of item, these are also one of the reason for strapdown inertial error dissipates；In addition, current strap down inertial navigation is defended Star integrated navigation system is not utilized sufficiently to information, is only just abandoned using once rear in filtered time instant.
The content of the invention
In view of this, the present invention provides a kind of Combinated navigation method and system, such that it is able to improve inertial navigation navigation accuracy and Optimization filter effect.
What technical scheme was specifically realized in：
A kind of Combinated navigation method, the method includes step as described below：
The data of the data of the first navigation system and the second navigation system are carried out information fusion by A, the first computing unit, are obtained Integrated navigation data after to fusion；
The historical data and the history number of the second navigation system of B, the second computing unit according to the first navigation system of record According to estimation obtains gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise acoustic matrix and sight used Survey the estimate of noise battle array；Each estimate that estimation is obtained is evaluated, judges whether each estimate is more accurately Parameter；Estimate is fed back to by the first computing unit according to judged result, corresponding parameter is modified.
Preferably, first navigation system is strapdown inertial navigation system SINS, the second navigation system is satellite navigation System.
Preferably, the step A includes the steps：
A1, in the data of the first navigation system, angle increment is calculated according to gyro umber of pulse, and according to accelerometer pulse Number calculating speed increment；
A2, attitude, speed and position that subsequent time is calculated according to angle increment and speed increment；
The data of A3, attitude, speed and the position that basis is calculated and the second navigation system, are combined navigation letter Breath fusion, the integrated navigation data after being merged；
A4, the gyro umber of pulse and accelerometer umber of pulse that record the generation of each moment, record SINS are each The velocity location being calculated, each velocity location that record is obtained by satellite navigation system records filter value.
Preferably, the historical data and the historical data of the second navigation system of first navigation system according to record, Estimation obtains gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise acoustic matrix and observation used The estimate of noise battle array comprises the following steps：
B1, historical data and the historical data of the second navigation system according to the first navigation system for recording, set heredity The initial population that algorithm needs；
The historical data and the second navigation system of each individuality in B2, initial population according to the first navigation system of record Historical data calculated, obtain each individual respective speed, position and attitude, at the same also obtain each individuality filtering after Speed, position and attitude；
It is B3, individual for each, judge whether the individuality meets preconditioned；If at least one individual satisfaction is default Condition, then it is assumed that estimate successfully, then feed back to the first computing unit by the individual corresponding zero degree item, first order, noise battle array, right Corresponding parameter is modified；Required if there is multiple individual satisfactions, then selected the best individuality of fitness；
Meet preconditioned if none of individuality, then perform step B4；
B4, genetic manipulation is carried out, generation population of future generation.
Preferably, the step B1 includes the steps：
Currency to system noise acoustic matrix is encoded；
According to historical data, used to gyroscope constant value and proportional, accelerometer constant value and proportional, filtering is Each amount in system noise battle array and observation noise battle array, with 1 times of currency as steplength, currency is median, produces n Number；
All gyroscope constant values to be estimated system used with proportional, accelerometer constant value and proportional, filtering The n groups array of noise battle array and observation noise gust is into A matrixes；
A matrixes homogenize and obtains matrix population, using matrix population as the first of genetic algorithm Beginning population.
Preferably, the step B2 includes the steps：
N estimation subelement, the position r that will be obtained from the second navigation system are set_{g}It is denoted as standard value；Take estimation subelement Position r in the same time_{i}With the position r in the second navigation system_{g}Make the difference, obtain position error vector δ r_{i}；Whether filtering is restrained Judged；When convergence is filtered, take the site error of steadystate portion and calculate the error 2 norms at each moment；When filtering is sent out When dissipating, 50 last site errors of filtering point are taken, calculate the error 2 norms at each moment；By the site error 2 models Number scale is δ r_{nij}, wherein, j=1,2 ... 50；Seek δ r_{nij}Average value and second geometric moment, Er is designated as respectively_{i}And Sr_{i}；
The inertia system continuous item after each estimation subelement filter correction is updated in respective inertial navigation recursion, from Initial time starts to calculate, until current time；The position at the current time remembered is r_{i}', the position with the second navigation system Put r_{g}The 2 norms of simultaneously calculation error are made the difference, is designated as
To each estimation subelement, its fitness scalar fitness is calculated_{i}=k_{1}δr_{i}'+k_{21}Er_{i}+k_{22}Sr_{i}；Wherein, k_{1}、 k_{21}、k_{22}It is weights, fitness_{i}It is the fitness scalar for describing the individual i that ith is estimated subelement distribution；It is individual by n The fitness vector f itness of fitness scalar composition description population's fitness；
Minimum value in fitness vector is corresponding individual as the best individuality of adaptability：S_{opt}=min {fitness}。
Preferably, the step B4 includes the steps：
B41, the individuality in parent population population is carried out to select to obtain population S_population；
B42, crossover operation is carried out to the individuality in population S_population obtain population C_population；
B43, mutation operation is carried out to the individuality in population C_population obtain progeny population N_population.
Preferably, the step B41 includes：
The probability that setting optimum individual has 30% is chosen to the next generation, and the probability that suboptimum individuality has 20% is chosen to down A generation, the individuality of adaptability the 3rd has 15% probability to of future generation, and the individuality of adaptability the 4th has 10% probability to next Generation, remaining individuality has 25% probability to of future generation；
Produce the random number between n 0~1, carry out n selection, to selecting every time, if random number 0~0.3 it Between, then select optimum individual；If random number is between 0.3~0.5, selection suboptimum is individual, the like；If random number Between 0.75~1, then a selection individual random from remaining individuality is genetic to the next generation.
Preferably, the step B42 includes：
Preset crosspoint number c_num, crossover probability P_{c}；
Crossover probability is multiplied into number of individuals, is then rounded, obtain the individual amount cross_num for being intersected；
Cross_num individuality is randomly chosen from population S_population, c_num intersection is randomly determined Point, intersects twobytwo between individuality；
To each crosspoint, specific interleaved mode is：
Wherein a and b are initial values, and a' and b' is the value obtained after intersecting, and α and β is the random number between 0~1.
Preferably, the step B43 includes：
Preset mutation probability P_{m}；
Preset the variances sigma of variation；
Produce and the random number between gene number gen equal number of 0~1；
If ith random number is less than mutation probability P_{m}, then ith gene enter row variation；If ith random number is more than Mutation probability P_{m}, then ith gene is not operated；Wherein, i takes 1,2,3...gen；
Specifically variation mode is：
C'=N (c, σ)；
Wherein c is initial value, and c' is the value obtained after making a variation.It is expectation that c' is taken with c, and variance is σ^{2}A random number.
Preferably, first navigation system is strapdown inertial navigation system SINS；
Second navigation system is earth magnetism integrated navigation system or scene matching aided navigation integrated navigation system.
Present invention also offers a kind of integrated navigation system, the system includes：
First computing unit and the second computing unit；
First computing unit, for the data of the data of the first navigation system and the second navigation system to be entered into row information Fusion, the integrated navigation data after being merged；
Second computing unit, for the historical data and the second navigation system of the first navigation system according to record Historical data, estimation obtains gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise used The estimate of battle array and observation noise battle array；Each estimate that estimation is obtained is evaluated, judges whether each estimate is more Accurate parameter；Estimate is fed back to by the first computing unit according to judged result, corresponding parameter is modified.
Preferably, first computing unit is further included：First data processing module, the second data processing module, Resolve module and filtration module；
First data processing module, angle increment and speed are calculated for the data according to the first navigation system for receiving Degree increment, and angle increment and speed increment are sent to resolving module；
Second data processing module, for the data is activation of the second navigation system that will receive to filtration module；
The resolving module, attitude, speed and position for calculating subsequent time according to angle increment and speed increment, and Attitude, speed and the position that will be calculated are sent to filtration module；
The filtration module, for according to the data for receiving, being combined navigation information fusion, the group after being merged Close navigation data.
Preferably, second computing unit is further included：Computing module, evaluation module and data memory module；
The data memory module, for storing the historical data of the first navigation system and the history number of the second navigation system According to；
The computing module, for the historical data and the history of the second navigation system of the first navigation system according to record Data, estimation obtain gyroscope constant value and proportional, accelerometer constant value and proportional, filtering system noise acoustic matrix used and The estimate of observation noise battle array；
The evaluation module, for evaluating each estimate for obtaining of estimation, judge each estimate whether be More accurately parameter；Estimate is fed back to by the first computing unit according to judged result, corresponding parameter is modified.
Preferably, the computing module is further included：Population operation subelement, resolving subelement, outcome evaluation are single Unit and n estimation subelement；
The population operates subelement, for the historical data and the second navigation system of the first navigation system according to record Historical data, set genetic algorithm need initial population, and according to outcome evaluation subelement return data carry out heredity Operation, generation population of future generation；
The resolving subelement, attitude, speed and position for calculating subsequent time according to angle increment and speed increment, And attitude, speed and the position that will be calculated are sent to n estimation subelement；
Described n is estimated that subelement corresponds respectively to n in initial population individuality, for the first navigation according to record The historical data of the historical data of system and the second navigation system is calculated, obtain each individual respective speed, position and Attitude, while also obtaining each individual filtered speed, position and attitude；
The outcome evaluation subelement, for individual for each, judges whether the individuality meets preconditioned；If At least one individuality meets preconditioned, then it is assumed that estimate successfully, then by the individual corresponding zero degree item, first order, noise battle array The first computing unit is fed back to, corresponding parameter is modified；Required if there is multiple individual satisfactions, then selected suitable The best individuality of response；Meet preconditioned if none of individuality, then return to the data that n is estimated subelement output The population operates subelement.
Preferably, the data storage of the first navigation system that first data processing module will be received is in data storage In module；
The data storage of the second navigation system that second data processing module will be received is in data memory module.
Preferably, the first guidance system data processing module is strapdown inertial navigation system data processing module；
The second guidance system data processing module is gps data processing module.
As above it is visible, in Combinated navigation method and system in the present invention, due to that can be navigated according to the first of record The historical data of the historical data of system and the second navigation system, estimation obtains gyroscope constant value and proportional, accelerometer are normal The estimate of system noise acoustic matrix and observation noise battle array used by value item and proportional, filtering；Each estimate obtained to estimation Evaluated, judged whether each estimate is more accurately parameter；Estimate is fed back to by the first navigation according to judged result System and the second navigation system, are modified to corresponding parameter, therefore can realize the integrated navigation based on SINS, and And the historical data information of two navigation system (for example, SINS and GPS) is more fully make use of, believe by historical data The excavation of breath and recycling, it is possible to achieve to gyro and the error compensation of accelerometer, realize to filtering system noise covariance The optimal designaside of battle array and observation noise covariance matrix, realize gyroscope constant value drift and proportional, accelerometer constant value drift and Proportional, the Online Estimation of Filtering Model system equation noise battle array and Filtering Model observational equation simultaneously feed back to corresponding system, Online amendment in real time can be carried out to the instrumental error of gyro and accelerometer, even therefore when there is no GPS information, also may be used To cause that SINS has precision higher during than without feedback, the purpose for improving inertial navigation navigation accuracy is reached.
In addition, in the inventive solutions, system equation noise battle array and observational equation also further to filtering Noise battle array has carried out online amendment in real time, reduces requirement when designing wave filter, improves noise covariance battle array description phase The accuracy of noise is answered, so as to improve filtering accuracy.Diverging to filtering also has certain inhibition, is more suitable for for a long time Filtering.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the Combinated navigation method in the embodiment of the present invention.
Fig. 2 is the idiographic flow schematic diagram of the step 11 in the embodiment of the present invention.
Fig. 3 is the idiographic flow schematic diagram of the step 12 in the embodiment of the present invention.
Fig. 4 is the structural representation of the integrated navigation system in the embodiment of the present invention.
Specific embodiment
To make the objects, technical solutions and advantages of the present invention become more apparent, develop simultaneously embodiment referring to the drawings, right The present invention is further described.
Present embodiments provide a kind of Combinated navigation method and system.
Fig. 1 is the schematic flow sheet of the Combinated navigation method in the embodiment of the present invention.As shown in figure 1, the embodiment of the present invention In Combinated navigation method mainly include step as described below：
The data of the data of the first navigation system and the second navigation system are entered row information and melted by step 11, the first computing unit Close, the integrated navigation data after being merged.
In the inventive solutions, two independent computing units can essentially be set, parallel computation is carried out, such as Shown in Fig. 2.Wherein, the function of first computing unit is to carry out conventional integrated navigation, i.e., used in current engineering Integrated navigation mode, the data of the data of the first navigation system (for example, SINS) and the second navigation system (for example, GPS) are entered Row information is merged, the integrated navigation data (performing abovementioned step 11) after being merged；And second work(of computing unit Can be then that required parameter (such as zero degree of accelerometer) is carried out estimating etc. to operate (performing aftermentioned step 12). Wherein, the first data processing module in first computing unit is used to process the data of the first navigation system, at the second data Reason module is then used to process the data of the second navigation system.
Therefore, in the inventive solutions, first computing unit can be seen as an integrated navigation system, And see second computing unit as a secondary navigation system.The integrated navigation system and secondary navigation system can be parallel Work；And, after secondary navigation system is damaged or loses function, integrated navigation system remains to be normally carried out work.
Preferably, in a particular embodiment of the present invention, first navigation system is SINS, the second navigation system is to defend Star navigation system (for example, gps system), or other navigation system.
Therefore, in this step, SINS data and gps data can be carried out information fusion, the combination after being merged Navigation data, and guarantee information renewal rate.
In the inventive solutions, abovementioned step 11 can be realized by various concrete implementation modes.With Under technical scheme will in detail be introduced by taking a kind of specific implementation therein as an example.It is specific at this In example, the first navigation system is SINS, and the second navigation system is satellite navigation system (for example, gps system).Second navigation system Unite as the situation of other navigation system so that the rest may be inferred, therefore can be repeated no more.
For example, preferably, Fig. 2 be the embodiment of the present invention in step 11 idiographic flow schematic diagram, as shown in Fig. 2 In specific embodiment of the invention, the step 11 can include step as described below：
Step 111, in the data of the first navigation system (SINS), according to gyro umber of pulse calculating angle increment, and according to Accelerometer umber of pulse calculating speed increment.
For example, in a preferred embodiment of the invention, it is possible to use formula as described below comes according to gyro Umber of pulse calculates angle increment：
Wherein, D_{0x}、D_{0y}、D_{0z}It is gyroscope constant value error coefficient, unit：Rad/the second；D_{xx}、D_{xy}、D_{xz}、D_{yx}、D_{yy}、D_{yz}、 D_{zx}、D_{zy}、D_{zz}It is gyroscope constant value error coefficient, unit：1；N_{gx}、N_{gy}、N_{gz}For gyro exports pulse number, k_{g}For gyro is worked as Amount；T is the sampling period.
Again for example, in a preferred embodiment of the invention, it is possible to use formula as described below carrys out basis and adds Speedometer umber of pulse calculating speed increment：
Wherein, E_{0x}、E_{0y}、E_{0z}It is the zero degree error coefficient of accelerometer, unit is g_{0z}；E_{xx}、E_{xy}、E_{xz}、E_{yx}、E_{yy}、 E_{yz}、E_{zx}、E_{zy}、E_{zz}It is the first order error coefficient of accelerometer, unit is g_{0z}；ΔN_{ax}、ΔN_{ay}、ΔN_{az}It is accelerometer Output pulse number；k_{a}To add meter equivalent；g_{0z}It is the relevant parameter of accelerometer, represents an acceleration of gravity list for standard Position, unit：m/s^{2}；T is the sampling period.
Step 112, attitude, speed and the position of subsequent time are calculated according to angle increment and speed increment.
For example, in a preferred embodiment of the invention, it is possible to use step as described below calculates lower a period of time The attitude at quarter, speed and position：
Step 112a, the attitude of subsequent time is calculated according to angle increment；
In this step, the attitude of subsequent time will be calculated according to angle increment first.
For example, in a preferred embodiment of the invention, it is possible to use formula as described below is next to calculate The attitude at moment：
Wherein,q_{k} It is the attitude quaternion at current time, q_{k+1}It is the attitude quaternion of subsequent time.
In addition, in the inventive solutions, attitude quaternion q_{k}With the Direct cosine matrix for representing attitudeBetween can Mutually to change and be correspond, subsequently will not be described again.
Step 112b, speed and the position of subsequent time are calculated according to speed increment.
For example, in a preferred embodiment of the invention, apparent velocity increment is changed in the value of carrier system through attitude It is the apparent velocity increment of navigational coordinate system, is multiplied by calculating cycle and the speed of navigational coordinate system is obtained plus initial value, to speed It is integrated and obtains position.
Because the navigational coordinate system used by different carriers is not quite similar, and the air navigation aid belongs to the known skill of this area Art, therefore here is omitted.In one of the invention specific preferred embodiment, the navigational coordinate system for being used can be hair Penetrate inertial system.
By abovementioned step 112a and 112b, you can be calculated the appearance of subsequent time according to angle increment and speed increment State, speed and position.
Step 113, according to the attitude, speed that are calculated and the data of position and the second navigation system, is combined Navigation information is merged, the integrated navigation data after being merged.
For example, in a preferred embodiment of the invention, it is possible to use kalman filter method is combined leads Boat information fusion, the integrated navigation data after being merged.
For example, in a preferred embodiment of the invention, abovementioned step 113 can include as described below The step of：
Step 113a, sets up the system equation and observational equation of continuous system；
For example, in a preferred embodiment of the invention, the system equation and observational equation can be expressed as：
Z (t)=H (t) X (t)+v (t) (5)
Wherein,
X (t)=(δ V_{x},δV_{y},δV_{z},δr_{x},δr_{y},δr_{z},δφ_{x},δφ_{y},δφ_{z},δK_{0x},δK_{0y},δK_{0z},δD_{0x},δD_{0y},δD_{0z} )^{T} (6)
Z (t)=(δ V_{x},δV_{y},δV_{z},δr_{x},δr_{y},δr_{z})^{T} (8)
V (t)=(Δ V_{gx},ΔV_{gy},ΔV_{gz},Δr_{gx},Δr_{gy},Δr_{gz})^{T} (9)
Wherein, δ V in state vector X (t)_{x},δV_{y},δV_{z}It is velocity error, δ r_{x},δr_{y},δr_{z}It is site error, δ φ_{x},δ φ_{y},δφ_{z}It is attitude error, δ K_{0x},δK_{0y},δK_{0z}It is the acceleration error of zero, δ D_{0x},δD_{0y},δD_{0z}It is gyroscope constant value error； In system white noise w (t)It is the noise of accelerometer, ε_{x1},ε_{y1},ε_{z1}It is the noise of gyro；Observation vector Z (t) Middle δ V_{x},δV_{y},δV_{z}It is the difference of the speed that inertia computing speed and satellite are measured, δ r_{x},δr_{y},δr_{z}For inertia resolves position and satellite The difference of the position for measuring；Δ V in observation white noise v (t)_{gx},ΔV_{gy},ΔV_{gz}It is velocity accuracy error, Δ r_{gx},Δr_{gy},Δr_{gz} It is positional precision error；I in sytem matrix F (t)_{3×3}It is unit matrix,The Direct cosine matrix of inertial system, f are tied to for carrier^{i} × for apparent acceleration in inertial system value composition antisymmetric matrix；System noise is shifted in battle array G (t) and observing matrix H (t) Element implication is ibid.
Step 113b, discretization is carried out to system equation and observational equation.
For example, in a preferred embodiment of the invention, after carrying out discretization for abovementioned equation (4), (5) Obtain：
X_{k+1}=Φ_{k+1,k}X_{k}+Γ_{k}w_{k} (13)
Z_{k+1}=H_{k+1}X_{k+1}+v_{k+1} (14)
Wherein, state transfer matrixInput matrix System noise covariance matrix Observation noise covariance matrixT is filtering cycle, F_{k}≈F (t),G_{k}≈ G (t), q are the variance intensity battle array of w (t), H_{k+1}=H (t), v_{k+1}=v (t).
Step 113c, is filtered to the system equation and observational equation after discretization.
For example, in a preferred embodiment of the invention, be filtered for abovementioned equation (13), (14), Obtain：
P_{k+1,k+1}=P_{k+1,k}K_{k+1}H_{k+1}P_{k+1,k} (19)
By abovementioned step 113a~113c, you can be combined navigation information fusion, the combination after being merged is led Boat data.
By abovementioned step 111~113, you can SINS data and gps data are carried out into information fusion, after being merged Integrated navigation data, and guarantee information renewal rate.
Step 12, the historical data of first navigation system and the going through of second navigation system of second computing unit according to record History data, estimation obtains gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise acoustic matrix used With the estimate of observation noise battle array；Each estimate that estimation is obtained is evaluated, judges whether each estimate is more accurate True parameter；Estimate is fed back to (for example, the first data in the first computing unit by the first computing unit according to judged result Processing module and filtration module), corresponding parameter is modified.
Preferably, in a particular embodiment of the present invention, before the step 12, can further include：
Record the gyro umber of pulse Δ N of each moment generation_{gx},ΔN_{gy},ΔN_{gz}With accelerometer umber of pulse Δ N_{ax},Δ N_{ay},ΔN_{az}, the velocity location that SINS is calculated every time is recorded, record each obtained by satellite navigation system Velocity location, records filter value.
In this step, the second computing unit will navigate according to the historical data of the first navigation system for being recorded and second The historical data of system, to gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise used Battle array and observation noise battle array estimated, and whether estimate obtained by judging is more accurately parameter；Then, further according to judgement Estimate is fed back to the first computing unit by result, and corresponding parameter is modified.
In the inventive solutions, abovementioned step 12 can be realized by various concrete implementation modes.With Under technical scheme will in detail be introduced by taking a kind of specific implementation therein as an example.It is specific at this In example, the first navigation system is SINS, and the second navigation system is satellite navigation system (for example, gps system).
For example, preferably, Fig. 3 be the embodiment of the present invention in step 12 idiographic flow schematic diagram, as shown in figure 3, In specific embodiment of the invention, the historical data of the first navigation system according to record in the step 12 and second is navigated The historical data of system, estimates that obtain gyroscope constant value is with used by proportional, accelerometer constant value and proportional, filtering The estimate of system noise battle array and observation noise battle array can include step as described below：
Step 121, the historical data and the historical data of the second navigation system of the first navigation system according to record are set The initial population that genetic algorithm needs.
In the inventive solutions, it is possible to use genetic algorithm carries out abovementioned estimation.Therefore, in this step In, will the initial population that genetic algorithm needs be set first.
For example, specifically, in a preferred embodiment of the invention, the step 121 can include as described below The step of：
Step 121a, the currency to system noise acoustic matrix is encoded.
The main purpose for carrying out the operation is to carry out nondimensionalization treatment, and make all parameters numerically will not difference too Greatly.
Step 121b, according to historical data, to abovementioned gyroscope constant value and proportional, accelerometer constant value item and ratio Each amount in item, filtering system noise acoustic matrix used and observation noise battle array, with 1 times of currency as steplength, currency is Median, produces n numbers；
Step 121c, all gyroscope constant values to be estimated and proportional, accelerometer constant value item and proportional, filtering institute The n groups array of system noise acoustic matrix and observation noise battle array is into A matrixes；
Step 121d, to A matrixes homogenize and obtains matrix population, using matrix population as heredity The initial population of algorithm.
By abovementioned step 121a~121d, you can obtain the initial population of genetic algorithm.
After abovementioned initial population homogenization, the convergence rate of genetic algorithm can be to a certain extent improved, reduce iteration Number of times, so as to reduce amount of calculation, saves the calculating time.So homogenization can be as the optimization method of genetic algorithm.
In addition, the homogenization method used in the present invention can be using conventional uniform design, for example, Fang Kaitai is carried The uniform design of confession, specifically used uniform designs table can be found in Fang Kaitai's《Uniform design》, will not be repeated here.
Step 122, each individuality in initial population is led according to the historical data and second of the first navigation system of record The historical data of boat system is calculated, and obtains each individual respective speed, position and attitude, while it is individual also to obtain each Filtered speed, position and attitude.
In this step, can be by each individual (each in initial population (actually contains zero degree, proportional) Individuality can correspond to a following estimation subelement) using the historical data of storage, (the first navigation system for recording is gone through The historical data of history data and the second navigation system) calculated, so as to obtain each individual respective speed, position and appearance State, while also obtaining each individual filtered speed, position and attitude (noise battle array or cry association that each individuality filtering is used Variance matrix is also different).
For example, specifically, in a particular embodiment of the present invention, if there is n individuality in set initial population, Then can first set n and estimate subelement (n corresponded respectively in initial population is individual), each estimates that son is single It is first substantial all equivalent to a SINS/GPS pine combinations navigation system (the first computing unit i.e. shown in Fig. 4), all include Inertia recursion and filtering two parts；Then, it is each individuality S in estimating subelement allocation matrix population；Then, To all estimation subelements, using identical initial time and initial value；The initial value of position and speed is using in the same time from defending Position and speed that star navigation system (for example, gps system) is obtained；Then, each estimates subelement to the zero degree in quantity of state Item etc. is filtered amendment, obtains corresponding filter value；Afterwards, you can according to the historical data of record, carry out recursion and filtering Calculate, repeat filtering, record each speed V for estimating all moment generations of subelement_{i}With position r_{i}。
For example, preferably, in a particular embodiment of the present invention, the step 122 can include the steps：
Step 122a, sets n estimation subelement, the position r that will be obtained from satellite navigation system (for example, gps system)_{g} It is denoted as standard value；Take and estimate subelement position r in the same time_{i}With the position r in gps system_{g}Make the difference, obtain position error vector δ r_{i}；Judge whether filtering restrains；When convergence is filtered, take the site error of steadystate portion and calculate the mistake at each moment Difference 2 norms；When filtering divergence, 50 last site errors of filtering point are taken, calculate the error 2 norms at each moment； The site error 2 norms are designated as δ r_{nij}, wherein, j=1,2 ... 50；Seek δ r_{nij}Average value and second geometric moment, respectively It is designated as Er_{i}And Sr_{i}。
Step 122b, by the inertia system continuous item after each estimation subelement filter correction (for example, accelerometer zero degree Etc.), it is updated in respective inertial navigation recursion, calculated since initial time, until current time；The current time remembered Position be r_{i}', the position r with satellite navigation system (for example, gps system)_{g}The 2 norms of simultaneously calculation error are made the difference, is designated as
Step 122c, to each estimation subelement, calculates its fitness scalar fitness_{i}=k_{1}δr_{i}'+k_{21}Er_{i}+ k_{22}Sr_{i}；Wherein Er_{i}、Sr_{i}、δr_{i}' as described above, k_{1}、k_{21}、k_{22}It is weights, fitness_{i}It is that description estimates subelement point ith The fitness scalar of the individual i for matching somebody with somebody；By the n fitness vector of individual fitness scalar composition description population's fitness fitness。
Step 122d, the minimum value in fitness vector is corresponding individual as the best individuality of adaptability：S_{opt}= min{fitness}。
Step 123, it is individual for each, judge whether the individuality meets preconditioned；If at least one individuality is full Foot is preconditioned, then it is assumed that estimate successfully, then by the individual corresponding zero degree (for example, including gyroscope constant value and acceleration Meter constant value), first order (for example, including gyro proportional and accelerometer proportional), noise gust is (for example, including system noise Acoustic matrix and observation noise battle array) the first computing unit is fed back to, corresponding parameter is modified；If there is multiple individual satisfactions Required, then selected the best individuality of fitness；
Meet preconditioned if none of individuality, then perform step 124；
That is, in this step, in will determine that whether the velocity location in each individuality has saltus step, estimating subelement Filtering whether restrain, system mode covariance matrix whether positive definite；The velocity potential of the second navigation system (for example, GPS) that will be stored Put as benchmark, made the difference with the velocity location for estimating to be obtained in subelement, judged according to default determination methods, if deposited Meet default all conditions at least one individuality, then it is assumed that estimate (i.e. step 12) success, return to corresponding zero degree, one Secondary item, noise battle array are (for example, zero degree therein and first order to be fed back to the first data processing mould in the first computing unit Block, and noise battle array is fed back to the filtration module in the first computing unit), required if there is multiple individual satisfactions, then The best individuality of selection fitness.Meet preconditioned if none of individuality, then perform step 124.
Wherein, the default determination methods are：
The fitness function of genetic algorithm is：
f_{1}(Q,P_{0})=k_{11}E(δV_{f}_{2})+k_{12}E(δr_{f}_{2})+k_{13}D(δV_{f}_{2})+k_{14}D(δr_{f}_{2})
Wherein, k_{1i}It is weight coefficient, i=1,2,3,4.
The optimization aim of genetic algorithm is to find the individuality for making above formula obtain minimum：
Wherein, Ω is the search space of genetic algorithm.
The variance of observation is designated asThe average value of velocity error time series is exp (  δ V  _{2}), during site error Between sequence average value be exp (  δ r  _{2}).Be description velocity error and site error dispersion degree, and exclude filtering can Can occur there is inclined situation, take its second geometric moment as description indexes, cov (  δ V   are designated as respectively_{2}) and cov (  δ r  _{2})。
The index of restrictive condition can be written as function：
f_{2}(Q)=k_{21}exp(δV_{2})+k_{22}exp(δr_{2})+k_{23}cov(δV_{2})+k_{24}cov(δr_{2})
Wherein, k_{2i}With k_{1i}Correlation, i=1,2,3,4.
For limitation Q diminishes extremely, constraints should be met：
J≥f_{2}(Q)
For having summarized, it should which it is sufficiently small to meet f1, while meeting f1 more than f2.
Step 124, carries out genetic manipulation, generation population of future generation.
In this step, genetic manipulation, generation population of future generation will be carried out.
For example, preferably, in a particular embodiment of the present invention, the step 124 can include the steps：
Step 124a, carries out selecting to obtain population S_population to the individuality in parent population population.
For example, preferably, in one particular embodiment of the present invention, can set optimum individual has 30% probability quilt The next generation is chosen, the probability that suboptimum individuality has 20% is chosen to the next generation, and the individuality of adaptability the 3rd has 15% probability To the next generation, the individuality of adaptability the 4th has 10% probability to of future generation, and remaining individuality has 25% probability to the next generation；So Afterwards, the random number between n 0~1 is produced, n selection is carried out, to selection every time, if random number is between 0~0.3, is selected Select optimum individual；If random number is between 0.3~0.5, selection suboptimum is individual, the like；If random number is 0.75 Between~1, then a selection individual random from remaining individuality is genetic to the next generation.
Step 124b, carries out crossover operation and obtains population C_population to the individuality in population S_population.
Further, in the present invention preferably specific embodiment, crosspoint number c_num can be preset, is handed over Fork probability P_{c}；Crossover probability is multiplied into number of individuals, is then rounded, obtain the individual amount cross_num for being intersected；Randomly from Cross_num individuality is chosen in population S_population, c_num crosspoint is randomly determined, handed over twobytwo between individuality Fork；To each crosspoint, specific interleaved mode is：
Wherein a and b are initial values, and a' and b' is the value obtained after intersecting, and α and β is the random number between 0~1.
Step 124c, carries out mutation operation and obtains progeny population N_ to the individuality in population C_population population。
Further, in the present invention preferably specific embodiment, mutation probability P can be preset_{m}；Due to being real Number encoder, can preset the variances sigma of variation；Then produce and the random number between gene number gen equal number of 0~1； If ith random number is less than mutation probability P_{m}, then ith gene enter row variation；If ith random number is more than mutation probability P_{m}, then ith gene is not operated；Wherein, i takes 1,2,3...gen；Specifically variation mode is：
C'=N (c, σ) (21)
Wherein c is initial value, and c' is the value obtained after making a variation.It is expectation that c' is taken with c, and variance is σ^{2}A random number.
By abovementioned step 124a~124c, you can carry out genetic manipulation, generation population of future generation.
In the description above, it with the first navigation system is SINS to be, the second navigation system is for satellite navigation system The introduction that example is carried out.However, in the inventive solutions, it is also possible to be as the second navigation from other navigation system System.For example, second navigation system can be：Earth magnetism integrated navigation system or scene matching aided navigation integrated navigation system etc. other lead Boat system.At this point it is possible to accordingly information fusion mode is changed accordingly, equally can be to some of fusion process Parameter is estimated.
In addition, in the inventive solutions, in addition to abovementioned described genetic algorithm, other can also be used Method of estimation, for example, preferably, in a particular embodiment of the present invention, other intelligence such as particle cluster algorithm can also be used Method will not be repeated here as method of estimation.
In addition, in the inventive solutions, two independent computing units can be set, parallel computation is carried out.Its In, the function of first computing unit is to carry out conventional integrated navigation, i.e., the integrated navigation side used in current engineering The data of the data of the first navigation system and the second navigation system are carried out information fusion by formula, the integrated navigation after being merged Data；And second function of computing unit is then that required parameter (such as zero degree of accelerometer) is estimated (performing abovementioned step 12).
Additionally, in the inventive solutions, data storage cell can also be set, navigated for online storage first Related data in system and the second navigation system, for example, the umber of pulse that the gyro and accelerometer in SINS are produced, GPS systems Speed position information obtained in system etc..
A kind of integrated navigation system is also proposed in the present invention.Fig. 4 is the integrated navigation system in the embodiment of the present invention Structural representation, as shown in figure 4, the integrated navigation system in the embodiment of the present invention includes：First computing unit and second is calculated Unit；
First computing unit, for the data of the data of the first navigation system and the second navigation system to be entered into row information Fusion, the integrated navigation data after being merged；
Second computing unit, for the historical data and the second navigation system of the first navigation system according to record Historical data, estimation obtains gyroscope constant value and proportional, accelerometer constant value item and proportional, filters system noise used The estimate of battle array and observation noise battle array；Each estimate that estimation is obtained is evaluated, judges whether each estimate is more Accurate parameter；Estimate is fed back to by the first computing unit according to judged result, corresponding parameter is modified.
Further, preferably, in a particular embodiment of the present invention, first computing unit can also be further Including：First data processing module, the second data processing module, resolving module and filtration module；
First data processing module, the data of the first navigation system received for basis are (for example, the top of SINS Spiral shell umber of pulse and accelerometer umber of pulse) angle increment and speed increment are calculated, and angle increment and speed increment are sent to resolving Module；
Second data processing module, the data of the second navigation system for that will receive are (for example, gps system Position and speed data) it is sent to filtration module；
The resolving module, attitude, speed and position for calculating subsequent time according to angle increment and speed increment, and Attitude, speed and the position that will be calculated are sent to filtration module；
The filtration module, for according to the data for receiving, being combined navigation information fusion, the group after being merged Close navigation data.
In addition, preferably, in a particular embodiment of the present invention, second computing unit can further include： Computing module, evaluation module and data memory module；
The data memory module, for storing the historical data of the first navigation system and the history number of the second navigation system According to；
The computing module, for the historical data and the history of the second navigation system of the first navigation system according to record Data, estimation obtain gyroscope constant value and proportional, accelerometer constant value and proportional, filtering system noise acoustic matrix used and The estimate of observation noise battle array；
The evaluation module, for evaluating each estimate for obtaining of estimation, judge each estimate whether be More accurately parameter；Estimate is fed back to (for example, in the first computing unit first by the first computing unit according to judged result Data processing module and filtration module), corresponding parameter is modified.
Additionally, preferably, in one particular embodiment of the present invention, the computing module can further include： Population operation subelement, resolving subelement, outcome evaluation subelement and n estimation subelement；
The population operates subelement, for the historical data and the second navigation system of the first navigation system according to record Historical data, set genetic algorithm need initial population, and according to outcome evaluation subelement return data carry out heredity Operation, generation population of future generation；
The resolving subelement, attitude, speed and position for calculating subsequent time according to angle increment and speed increment, And attitude, speed and the position that will be calculated are sent to n estimation subelement；
Described n is estimated that subelement corresponds respectively to n in initial population individuality, for the first navigation according to record The historical data of the historical data of system and the second navigation system is calculated, obtain each individual respective speed, position and Attitude, while also obtaining each individual filtered speed, position and attitude；
The outcome evaluation subelement, for individual for each, judges whether the individuality meets preconditioned；If At least one individuality meets preconditioned, then it is assumed that estimate successfully, then by the individual corresponding zero degree item, first order, noise battle array The first computing unit is fed back to, corresponding parameter is modified；Required if there is multiple individual satisfactions, then selected suitable The best individuality of response；Meet preconditioned if none of individuality, then return to the data that n is estimated subelement output The population operates subelement.
Further, in a preferred embodiment of the invention, first data processing module will can also be received To the first navigation system data storage in data memory module；
The data storage of the second navigation system that second data processing module will can also be received is in data storage mould In block.
Additionally, in a preferred embodiment of the invention, the first guidance system data processing module can be SINS data processing modules, the second guidance system data processing module can be gps data processing module, or other lead The data processing module of boat system.
In summary, in Combinated navigation method and system in the present invention, due to that can be navigated according to the first of record The historical data of the historical data of system and the second navigation system, estimation obtains gyroscope constant value and proportional, accelerometer are normal The estimate of system noise acoustic matrix and observation noise battle array used by value item and proportional, filtering；Each estimate obtained to estimation Evaluated, judged whether each estimate is more accurately parameter；Estimate is fed back to by the first navigation according to judged result System and the second navigation system, are modified to corresponding parameter, therefore can realize the integrated navigation based on SINS, and And the historical data information of two navigation system (for example, SINS and GPS) is more fully make use of, believe by historical data The excavation of breath and recycling, it is possible to achieve to gyro and the error compensation of accelerometer, realize to filtering system noise covariance The optimal designaside of battle array and observation noise covariance matrix, realize gyroscope constant value drift and proportional, accelerometer constant value drift and Proportional, the Online Estimation of Filtering Model system equation noise battle array and Filtering Model observational equation simultaneously feed back to corresponding system, Online amendment in real time can be carried out to the instrumental error of gyro and accelerometer, even therefore when there is no GPS information, also may be used To cause that SINS has precision higher during than without feedback, the purpose for improving inertial navigation navigation accuracy is reached.
In addition, in the inventive solutions, system equation noise battle array and observational equation also further to filtering Noise battle array has carried out online amendment in real time, reduces requirement when designing wave filter, improves noise covariance battle array description phase The accuracy of noise is answered, so as to improve filtering accuracy.Diverging to filtering also has certain inhibition, is more suitable for for a long time Filtering.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all in essence of the invention Within god and principle, any modification, equivalent substitution and improvements done etc. should be included within the scope of protection of the invention.
Claims (10)
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

CN201710005473.4A CN106767788B (en)  20170104  20170104  A kind of Combinated navigation method and system 
Applications Claiming Priority (1)
Application Number  Priority Date  Filing Date  Title 

CN201710005473.4A CN106767788B (en)  20170104  20170104  A kind of Combinated navigation method and system 
Publications (2)
Publication Number  Publication Date 

CN106767788A true CN106767788A (en)  20170531 
CN106767788B CN106767788B (en)  20190719 
Family
ID=58950136
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CN201710005473.4A CN106767788B (en)  20170104  20170104  A kind of Combinated navigation method and system 
Country Status (1)
Country  Link 

CN (1)  CN106767788B (en) 
Cited By (1)
Publication number  Priority date  Publication date  Assignee  Title 

CN107462904A (en) *  20170728  20171212  东南大学  The highprecision GNSS terminal dynamic detection and localization car and detection method of GNSS and INS fusions 
Citations (9)
Publication number  Priority date  Publication date  Assignee  Title 

CN101038170A (en) *  20070207  20070919  北京航空航天大学  Method for online estimating inertia/satellite combined guidance system data asynchronous time 
CN101949703A (en) *  20100908  20110119  北京航空航天大学  Strapdown inertial/satellite combined navigation filtering method 
CN103278837A (en) *  20130517  20130904  南京理工大学  Adaptive filteringbased SINS/GNSS (strapdown inertial navigation system/global navigation satellite system) multistage faulttolerant integrated navigation method 
CN103344259A (en) *  20130711  20131009  北京航空航天大学  Method for correcting feedback of inertial navigation system/global position system (INS/GPS) combined navigation system based on lever arm estimation 
CN103776449A (en) *  20140226  20140507  北京空间飞行器总体设计部  Moving base initial alignment method for improving robustness 
CN103968843A (en) *  20140521  20140806  哈尔滨工程大学  Selfadaption mixed filtering method of GPS/SINS (Global Positioning System/Strapdown Inertial Navigation System) supercompact integrated navigation system 
CN103994763A (en) *  20140521  20140820  北京航空航天大学  SINS (Ship's Inertial Navigation System)/CNS (Celestial Navigation System) deep integrated navigation system of mar rover, and realization method of system 
CN105758401A (en) *  20160514  20160713  中卫物联成都科技有限公司  Integrated navigation method and equipment based on multisource information fusion 
CN106199668A (en) *  20160624  20161207  南京理工大学  A kind of tandem type GNSS/SINS deep integrated navigation method 

2017
 20170104 CN CN201710005473.4A patent/CN106767788B/en active IP Right Grant
Patent Citations (9)
Publication number  Priority date  Publication date  Assignee  Title 

CN101038170A (en) *  20070207  20070919  北京航空航天大学  Method for online estimating inertia/satellite combined guidance system data asynchronous time 
CN101949703A (en) *  20100908  20110119  北京航空航天大学  Strapdown inertial/satellite combined navigation filtering method 
CN103278837A (en) *  20130517  20130904  南京理工大学  Adaptive filteringbased SINS/GNSS (strapdown inertial navigation system/global navigation satellite system) multistage faulttolerant integrated navigation method 
CN103344259A (en) *  20130711  20131009  北京航空航天大学  Method for correcting feedback of inertial navigation system/global position system (INS/GPS) combined navigation system based on lever arm estimation 
CN103776449A (en) *  20140226  20140507  北京空间飞行器总体设计部  Moving base initial alignment method for improving robustness 
CN103968843A (en) *  20140521  20140806  哈尔滨工程大学  Selfadaption mixed filtering method of GPS/SINS (Global Positioning System/Strapdown Inertial Navigation System) supercompact integrated navigation system 
CN103994763A (en) *  20140521  20140820  北京航空航天大学  SINS (Ship's Inertial Navigation System)/CNS (Celestial Navigation System) deep integrated navigation system of mar rover, and realization method of system 
CN105758401A (en) *  20160514  20160713  中卫物联成都科技有限公司  Integrated navigation method and equipment based on multisource information fusion 
CN106199668A (en) *  20160624  20161207  南京理工大学  A kind of tandem type GNSS/SINS deep integrated navigation method 
Cited By (1)
Publication number  Priority date  Publication date  Assignee  Title 

CN107462904A (en) *  20170728  20171212  东南大学  The highprecision GNSS terminal dynamic detection and localization car and detection method of GNSS and INS fusions 
Also Published As
Publication number  Publication date 

CN106767788B (en)  20190719 
Similar Documents
Publication  Publication Date  Title 

Li et al.  Learning to generalize: Metalearning for domain generalization  
CN104736963B (en)  mapping system and method  
Hager  Taskdirected sensor fusion and planning: a computational approach  
Blackard et al.  Comparative accuracies of artificial neural networks and discriminant analysis in predicting forest cover types from cartographic variables  
Imsland et al.  Vehicle velocity estimation using nonlinear observers  
Angelov et al.  A new type of simplified fuzzy rulebased system  
McKelvey et al.  Quantal response equilibria for normal form games  
Neuts  Algorithmic Probability: A collection of problems  
Bishop et al.  An introduction to the kalman filter  
Harold Jeffreys  Scientific inference  
Gillison  Transects or Gradsects in  
Simon  Optimal state estimation: Kalman, H infinity, and nonlinear approaches  
EP0526969B1 (en)  Tactical route planning method for use in simulated tactical engagements  
Bevly et al.  Integrating INS sensors with GPS measurements for continuous estimation of vehicle sideslip, roll, and tire cornering stiffness  
Risi et al.  Evolving plastic neural networks with novelty search  
Rice et al.  Next generation marine precision navigation system  
Forrester  Nonlinearity in highorder models of social systems  
Montemerlo et al.  FastSLAM 1.0  
Walsh  The measurement of general exchangevalue  
JP5300739B2 (en)  Method for controlling and / or adjusting a technical system with computer assistance  
Camerer et al.  Models of thinking, learning, and teaching in games  
Filzmoser et al.  Testing hypotheses with fuzzy data: the fuzzy pvalue  
Kasprzak et al.  Pareto analysis in multiobjective optimization using the collinearity theorem and scaling method  
Mourikis et al.  Sckf mobile robot localization: A stochastic cloning kalman filter for processing relativestate measurements  
Callaghan et al.  Random walker ranking for NCAA division IA football 
Legal Events
Date  Code  Title  Description 

PB01  Publication  
PB01  Publication  
SE01  Entry into force of request for substantive examination  
SE01  Entry into force of request for substantive examination  
GR01  Patent grant  
GR01  Patent grant 